Fix compile fusion for multi-output edge cases (#950)

* Fix compile fusion for multi-output edge cases

* Add a test for multi-output compile

mlx/compile.cpp

@@ -165,25 +165,32 @@ CompileMode& compile_mode() {
 using ParentsMap =
     std::unordered_map<std::uintptr_t, std::vector<std::pair<array, int>>>;
 
+// Helper like below but only merges the two provided arrays. If the src has
+// siblings then these won't be merged to the dst.
+void merge_one(array& dst, array& src, ParentsMap& parents_map) {
+  auto src_parents = parents_map.find(src.id());
+  if (src_parents == parents_map.end()) {
+    return;
+  }
+  auto& pairs = parents_map[dst.id()];
+  for (auto& parent : src_parents->second) {
+    parent.first.inputs()[parent.second] = dst;
+    pairs.push_back(parent);
+  }
+  // Remove the source from the map to avoid fusing with it again
+  parents_map.erase(src_parents);
+};
+
 // Helper that merges two arrays in the graph by setting the parents of the
-// source to point to the destination
+// source to point to the destination. The arrays are assumed to be coming from
+// equivalent primitives so their siblings are merged as well.
 void merge(array& dst, array& src, ParentsMap& parents_map) {
   // Canonicalize the order of the primitives outputs
   auto sources = src.outputs();
   auto dests = dst.outputs();
   // For each src parent, point it to the corresponding dst
   for (int i = 0; i < sources.size(); ++i) {
-    auto src_parents = parents_map.find(sources[i].id());
-    if (src_parents == parents_map.end()) {
-      continue;
-    }
-    auto& pairs = parents_map[dests[i].id()];
-    for (auto& parent : src_parents->second) {
-      parent.first.inputs()[parent.second] = dests[i];
-      pairs.push_back(parent);
-    }
-    // Remove the source from the map to avoid fusing with it again
-    parents_map.erase(src_parents);
+    merge_one(dests[i], sources[i], parents_map);
   }
 };
 
@@ -524,9 +531,14 @@ void compile_fuse(
     //  - Collect inputs to the new compiled primitive
     //  - Add fusable primitives to a tape in the correct order
 
-    std::function<void(const array&, int, const Stream&)> recurse;
+    std::function<void(
+        const array&, int, const Stream&, const std::vector<int>&)>
+        recurse;
     std::unordered_set<uintptr_t> cache;
-    recurse = [&](const array& a, int depth, const Stream& s) {
+    recurse = [&](const array& a,
+                  int depth,
+                  const Stream& s,
+                  const std::vector<int>& shape) {
       if (cache.find(a.id()) != cache.end()) {
         return;
       }
@@ -536,8 +548,10 @@ void compile_fuse(
       // - Constant input
       // - Stream mismatch
       // - Non fusable primitive
+      // - Is global output but has a different shape
       if (depth >= max_compile_depth || !a.has_primitive() ||
-          a.primitive().stream() != s || !is_fusable(a.primitive())) {
+          a.primitive().stream() != s || !is_fusable(a.primitive()) ||
+          (output_map.find(a.id()) != output_map.end() && a.shape() != shape)) {
         return;
       }
 
@@ -564,13 +578,13 @@ void compile_fuse(
       cache.insert({a.id()});
 
       for (auto& in : a.inputs()) {
-        recurse(in, depth + 1, s);
+        recurse(in, depth + 1, s, shape);
       }
     };
 
     if (arr.has_primitive()) {
       Stream s = arr.primitive().stream();
-      recurse(arr, 0, s);
+      recurse(arr, 0, s, arr.shape());
     }
 
     // Not worth fusing a single primitive
@@ -634,6 +648,10 @@ void compile_fuse(
     std::vector<std::vector<int>> shapes;
     std::vector<Dtype> types;
     for (auto& o : old_outputs) {
+      if (o.shape() != old_outputs.back().shape()) {
+        throw std::runtime_error(
+            "[compile] Compilation failed. Tried to fuse operations with different output shapes");
+      }
       shapes.push_back(o.shape());
       types.push_back(o.dtype());
     }
@@ -676,7 +694,7 @@ void compile_fuse(
     // - Update outputs parents to point to compiled outputs
     // - Update any overall graph outputs to be compiled outputs
     for (int o = 0; o < old_outputs.size(); ++o) {
-      merge(compiled_outputs[o], old_outputs[o], parents_map);
+      merge_one(compiled_outputs[o], old_outputs[o], parents_map);
       if (auto it = output_map.find(old_outputs[o].id());
           it != output_map.end()) {
         it->second = compiled_outputs[o];

python/tests/test_compile.py

@@ -691,6 +691,19 @@ class TestCompile(mlx_tests.MLXTestCase):
         out = mx.compile(fn)(mx.array(10.0), mx.array(20.0))
         self.assertEqual(out.item(), 10.0)
 
+    def test_compile_multi_output(self):
+        def fn(x):
+            ys = [x]
+            for i in range(5):
+                ys.append(ys[-1] + x)
+            return ys, mx.sum(ys[-1])
+
+        x = mx.ones(1, dtype=mx.int32)
+        y1 = mx.compile(fn)(x)[1]
+        y2 = fn(x)[1]
+        self.assertEqual(y1.item(), y2.item())
+        self.assertEqual(y1.item(), 6)
+
 
 if __name__ == "__main__":
     unittest.main()